me

exclusion of the other two hypotheses chanical contact and arrest of growth- they confined themselves to the last one, sensitiveness to contact. This would have been a fine field for a discussion of the known facts followed by a necessary inference. Of the three possible hypotheses, two had been excluded, and the third must be true. It would seem quite clear that the case was logically proved. But instead of making this the end, they made it the beginning of their work.

They "thought that any small hard object affixed to the tip of a radicle freely suspended and growing in damp air, might cause it to bend if it were sensitive, and yet would not offer any mechanical resistance to its growth." The results of their experiments proved remarkable. When approaching the subject they made a preliminary trial with seven beans at a rather cool temperature, and six radicles curved. Το quote again, “These six striking cases almost convinced us that the apex was sensitive, but of course we determined to make many more trials." As they had noticed that radicles grew much more quickly when subjected to considerable heat, and as they imagined that heat would increase their sensitiveness, they made five or six dozen trials on more than two dozen

beans at a temperature of 69°-72° F. The result was moderately distinct deflection in only one radicle; in five other cases slight and doubtful deflection. "We were astonished at this result, and concluded that we had made some inexplicable mistake in the first six experiments. But before finally relinquishing the subject, we resolved to make one other trial, for it occurred to us that sensitiveness is easily affected by external conditions, and that radicles growing naturally in the earth in the early spring would not be subjected to a temperature nearly so high as 70° F." In the vast number of successful trials that they made they allowed the radicles to grow at a temperature of 55°-60° F.1

Had they stopped with the first trial, they would have hit the explanation which they finally adopted, and missed the effect of variations in temperature. Had they stopped with the second, the question would have hung in the balance between contradictory results. It was very feasible to reason that the results of the older experiment were due to some error of observation or manipulation. Nothing would have been known concerning the effect of temperature, and nothing concerning the original

1 Power of Movement in Plants, pp. 141, 142.

question. Their imagination had led them to introduce a new element into the second experiment; then reason prevented them from succumbing before the disturbance in the results, and led them to recognize it as a determining factor and treat it as such in their subsequent experiments.

"He

The contradictory state in which things would have been left at the end of their second experiment is neatly illustrated by another case in connection with the same subject.1 Ciesielski had shown, in his study of geotropic movements, that roots extending horizontally with their tips cut off did not grow downward. further states that, if the tips are cut off after the roots have been left extended horizontally for some little time, but before they have begun to bend downwards, they may be placed in any position and yet will bend as if still acted on by geotropism; and this shows that some influence had been already transmitted to the bending part from the tip before it was amputated.” Sachs repeated these experiments, but denied the conclusions, because in his experiments the roots became distorted in all directions. The Darwins undertook to learn the cause of the contradiction in the results. After describ

1 Power of Movement in Plants, p. 523.

ing unsuccessful efforts based on reasoning, they go on to say, "We next thought that, if care were not taken in cutting off the tips transversely, one side of the stump might be irritated more than the other, either at first, or subsequently during the regeneration of the tip, and that this might cause the radicle to bend to one side." They amputated some radicles obliquely and some transversely, and allowed them to grow perpendicularly. There was little or no distortion at first; but after two or three days, when the new tips began to form, the distortion of the obliquely amputated radicles became very conspicuous. The new tip was probably formed obliquely, causing the bending. Sachs probably "unintentionally amputated the radicles not strictly transversely"; and by not attending to this apparently insignificant condition he produced confusion and failed to make a discovery.

This case is interesting not only because it illustrates the difficulties that are met by the individual investigator, but because it is a typical example of a very large proportion of contradictions in results with which the literature. of science is burdened. The contradiction in the results obtained by the two men was due, not to errors of observation, but to neglect of

the various conditions under which the experiments were made. As I have elsewhere shown

for another more involved case, the disputant observers were both right.1 It was lack of exhaustion of the logical conditions of the problem that led to the contradiction. Total exclusion of error requires that every movement of the experimenter be fraught with intention. It is fairly safe to assume that, if two observers are competent and upright, their contradictory results, no matter on what subject, will prove essential to the final solution of the problem.

The publication of the "Power of Movement in Plants" was followed by several years of active investigation on and discussion of the "Darwinian curvature" of radicles. It has been shown that radicles, instead of being deflected by tin-foil on soft sand, will penetrate mercury and pierce tin-foil even when they strike it at a quite high angle. In explanation of the initial error of the Darwins it has been suggested that the radicles upon which they experimented were wilted. It has been further shown that in the experiments in which they attached small hard objects to the tips of the radicles to induce them to curve, the curva

1 Popular Science Monthly, January, 1894, pp. 373–376.